This proposal is to conduct genetic and epidemiologic studies of bipolar disorder (BP). Specifically, it is based on the investigation of large pedigrees in Costa Rica that are potentially exceptionally informative for linkage analysis. Costa Rican scientific, governmental, and health care institutions have provided enthusiastic cooperation and support to this project. Three extended pedigrees have been identified and are the focus of current and planned investigation. A collaborative group has been established and has conducted substantial preliminary evaluations of these pedigrees, including collection of genealogical information and family psychiatric histories, structured diagnostic interviewing, review of medical records, collection of blood samples and establishment of permanent cell lines, and use of DNA marker studies to confirm he identity of collected samples. Extensive training and checking procedures have been set up to ensure the reliability of diagnostic interviews in English and Spanish. Medical records are obtained from a centralized database at the National Psychiatric Hospital in Costa Rica. Best estimate diagnoses will be established by consensus of highly experienced clinicians using all available data. Cell lines will be established at UCSF. Genotyping studies will be performed using samples from informative individuals using highly polymorphic DNA markers. Linkage simulation studies will be performed to estimate the informativeness for linkage of each branch of all families. Simulations and actual linkage analyses will include standard lod score and non-parametric methods, as well as a novel approach (circuit analysis) developed for this project which permits investigation of pedigrees in the presence of genetic heterogeneity through subdivision of each family into a set of circuits. Complex segregation analyses will be used to estimate genetic parameters for these studies. The diagnostic data will be used to investigate possible cohort effects for affective illness in this population.